Conducting a gap analysis for a publication plan

A publication gap analysis compares existing coverage with the information a product's audiences need. It identifies gaps in evidence, topics, audience reach and timing, then turns those findings into recommendations for studies and publications.

The questions a gap analysis answers

A gap analysis assesses how a drug, device, procedure or therapeutic area appears in medical literature and scientific congresses against internal goals, key topic points, future plans and competitors. ^{1, 2} It examines what information is available, where it appears, its format, its audiences (primary-care physicians, specialists, nurses and pharmacists) and timing, including unmet needs in the target indication related to competitors' claims. ³ A topic inadequately covered or failing to reach all appropriate audiences relative to internal standards or competitors is a knowledge gap, and an analysis may be repeated throughout a product's life. ^{3, 2}

One basic check is whether a primary manuscript is planned for every clinical trial, helping assess regulatory and moral obligations to communicate findings. ³ Timing matters as much as availability: information arriving after its audience needs it leaves as large a gap as not communicating it. ³

For example, a brand presenting only at specialist meetings may miss nurses and primary-care physicians reached by competitors; recommendations can include abstracts, posters, presentations and journal articles for those audiences. ³ Concentration in one journal category can likewise prompt a broader target list when competitors reach general medicine, cardiovascular, pharmacology, obesity, managed-care and health-technology audiences. ³

Analysis should seek transparent, balanced and timely information for healthcare professionals, rather than merely matching competitors' publication volume. ⁴ A producer lagging behind competitors remains responsible for making its evidence available so clinicians can make informed decisions on all available evidence. ⁴

Quantitative and qualitative assessment

Publication activity can be analysed quantitatively, qualitatively or through both approaches; a hybrid is generally most effective. ^{5, 6} Quantitative assessment counts publications about a product or device, alone or against competitors, using categories such as the following. ^{5, 4}

• Publication year or timeframe.
• Therapeutic indications covered.
• Keywords, such as population subgroup or efficacy endpoint.
• Publication types: journal reports of primary efficacy, safety or pharmacokinetic data, reviews, secondary analyses, editorials, letters, case reports, health-economic analyses, meta-analyses and guidelines; or congress abstracts, posters and presentations.
• Journal types, stratified by audience, specialty or geographic region.
• Mentions of key topic points, competitor claims and patient populations.
• Citations of selected articles during a defined period.

^{5, 4}

Three output examples are publication totals by article type, one product's citations over eight years, and publication counts for a product and competitors over eight years. ⁴ Absence of case reports and guideline coverage can prompt recommendations for more of those publications, while citation trends can inform steps to produce more citable work. ⁴

Qualitative analysis can compare positive or negative presentation of key topics and data, including differences in the language describing a product's evidence and competitors' evidence. ⁶ It can stratify publications by these categories. ⁶

• Data integrity or evidence strength based on trial-design quality.
• Exposure of key topic points.
• Exposure of key data.
• Inclusion in or exclusion from treatment guidelines.
• Developments in the therapeutic area.

⁶

Study-quality comparisons identify strong, weak or missing evidence; planned publications can emphasise strong evidence, while weak or absent evidence can prompt reevaluation and better study designs. ⁶ A product may have abundant coverage overall yet lag a competitor on one key topic, suggesting more publications on that topic. ⁶

Guideline analysis distinguishes absence caused by missing data from insufficient literature exposure, recent inclusion and inclusion below competitors' standing. ⁶ Expanded indications, new meta-analyses or population studies, and new findings on disease progression or pathogenesis can reveal coverage that needs updating or evidence requiring new studies. ^{6, 7} For example, a later paediatric indication can leave a product behind competitors in coverage, while a new meta-analysis can expose safety, efficacy, endpoint or design differences that publications should address. ⁷

Trial registries reveal competitors' study numbers and status, designs, established and evolving endpoints, indications and populations; completed, ongoing and planned trials indicate potential future publications. ⁴ Comparing planned trial endpoints with a product's key topic points shows whether the programme will generate relevant, comparably informative data and can identify additional endpoints or analyses needed. ^{4, 6}

The seven-step procedure

A gap analysis follows seven steps, with the first five preparing the search. ^{1, 8, 9}

1. Identify the focus and rationale. Define a limited topic question, informed by a comprehensive understanding of the therapeutic area and competitors. For example, compare one brand's postlaunch journal coverage with three competitors to ask which had the most activity, what evidence was featured and which trial endpoints were common.
2. Determine a meaningful timeline. Choose a historical period, a current window or a milestone snapshot. Historical analysis can benchmark lifecycle phases and identify trends for forecasting; a recent window such as 12 months shows current issues and permits deeper examination of less material. Milestone snapshots compare prelaunch, perilaunch or postlaunch periods to reveal trends.
3. Identify the research scope. Set publication types and databases before searching. Common materials are indexed peer-reviewed articles and meeting abstracts, posters and presentations. Choose databases such as PubMed, EMBASE, the Cochrane Library, Google Scholar, CINAHL, PsycINFO or Web of Knowledge, and decide whether abstracts qualify when full articles are unavailable. Preselect congresses and search organisations' archives. Other sources can include national trial registries, government, medical-organisation and disease-association websites; press releases can provide qualitative starting points, targets and audiences through press-release databases or general search engines.
4. Determine search parameters. Establish sources, keywords, filters and limits from the focus, timeframe and scope before research begins, then follow them rigidly for reproducibility and record inclusion and exclusion rationales. For a drug, begin with its name and relevant competitors, then consider disease, treatment type and data such as quality of life or efficacy. Limits may cover text availability, language, article type, population age and publication dates. Allow for several analyses, including product-specific, disease-specific and patient-related sets.
5. Select the output format. Decide how results will be presented and organised before searching. A word-processing table or spreadsheet grid can track data, with content imported into slides for presentation.
6. Conduct the search. Follow the established keywords strictly and populate the output carefully with publication-coverage data.
7. Organise and prioritise findings against the topic question. Match grid columns to search parameters, such as citation, journal or congress, title, publication type, drug, endpoints, key topics and key data. Add useful categories emerging from the findings, such as audience, journal type or meeting type; alphabetising columns helps organise results. Interpret trends across categories to formulate recommendations.

^{8, 10, 11, 9, 12, 13, 14, 15}

For example, change in glycated haemoglobin is common in publications about exenatide and liraglutide: a competitor using different endpoints can compare its trials and consider studies using that primary endpoint, followed by abstracts, posters and articles. ¹⁵

Trends and recommendations

The findings ideally describe the therapeutic area through its publication landscape, comparing a product with its previous performance or competitors to reveal evidence, theme and audience gaps. ¹⁶ Five areas help identify trends. ¹⁶

• Topics differentiating the product from competitors.
• Topics central to its role or identity in the therapeutic landscape.
• Common themes, including positive, negative or neutral depictions relative to competitors.
• Coverage of the product's and competitors' safety and efficacy.
• Coverage of specific symptoms and special populations.

¹⁶

A review can reveal the following patterns. ¹⁶

• Common trial endpoints.
• Journals or journal types targeted more often than others.
• A few recurring lead authors.
• Proportions of reviews, primary data reports and secondary analyses.
• Trials lacking primary manuscripts.

¹⁶

Recommendations may include endpoints for future trials according to lifecycle stage, using existing or forthcoming trial data to fill gaps, or seeking an expert's participation when that expert repeatedly authors competitors' publications. ¹⁶ The analysis identifies data to emphasise, audiences to reach, journals and congresses to target, and submission timing. ^{2, 12, 13, 14}

Gap analyses should follow publication-ethics guidance on author freedoms and responsibilities, authorship criteria, data reporting and conflict-of-interest disclosure. ^{1, 17, 2}